Diagnostic errors are common, deadly, and costly. Twelve million Americans annually experience diagnostic error in ambulatory care, including in Emergency Department (EDs), over half of these with potential for harm. ED clinical practice is especially prone to diagnostic error as a sociotechnical work system that is fast-paced, high-stakes, highly adaptive and complex. The 2016 National Academy of Medicine (NAM) report was an urgent call for more research regarding diagnostic safety, making particular reference to the ED. ED diagnosis is cognitively-intense work, distributed across team members who may or may not be co-located. There is very limited understanding of the salient `real-time' details of the ED diagnostic process and associated performance shaping factors on the work system. Without structured in-depth analysis of ED diagnosis occurring as part of `real-time ED work,' that is ?work-as-done,? we will continue the struggle with the design of effective, sustainable interventions to improve diagnostic safety. Accordingly, we are proposing a 3-year, multi- site, multi-method field study in the ED based on a sociotechnical systems approach and a macrocognition framework, which is the study of cognitive tasks that characterize how people think in natural settings. We have 3 specific aims: (1) AIM 1. To understand provider (physician and advanced practice provider) work involved in ED diagnosis and identify associated performance shaping factors. (2) AIM 2. To understand collaborative (team-oriented) work involved in ED diagnosis and identify associated performance shaping factors. (3) AIM 3. To conduct a proactive risk assessment of the diagnostic process in the ED. AIM 1 and AIM 2 will be achieved by conducting in-depth qualitative studies using a variety of data collection methods (observations, interviews) and cognitive task analyses techniques. Data analysis will produce a range of outputs such as process maps, macrocognitive and procedural tasks involved in diagnosis, information flow diagrams, role network graphs, among others. AIM 3 will use two complementary proactive risk assessment methods to assess failure modes and performance shaping factors and to identify possible interventions to improve ED diagnostic safety: (1) Health Care Failure Mode and Effect Analysis (HFMEA); (2) Functional Resonance Analysis Method (FRAM) Based ?What-if? Risk Analysis. Additionally, we will develop a research methods compendium/guide for those interested in conducting similar research on diagnostic safety. The study will be conducted in 3 different EDs (urban, suburban, rural) that serve patients from 6 AHRQ priority population groups. The research team is interdisciplinary, composed of internationally known experts in patient safety, human factors, systems engineering, cognitive psychology, communication, emergency medicine, and nursing. The study is innovative due to its lens on ED diagnostic process as a whole, its use of human factors- based conceptual approaches, its investigation of the ED team's role in the diagnosis, and its use of a variety of cognitive task analysis techniques and proactive risk assessment methods.